Hubbry Logo
Full RateFull RateMain
Open search
Full Rate
Community hub
Full Rate
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Contribute something
Full Rate
Full Rate
Full Rate
View on Wikipedia
from Wikipedia
Audio quality comparison
Full Rate
Original file
Problems playing these files? See media help.

Full Rate (FR), also known as GSM-FR or GSM 06.10 (sometimes simply GSM), was the first digital speech coding standard used in the GSM digital mobile phone system. It uses linear predictive coding (LPC). The bit rate of the codec is 13 kbit/s, or 1.625 bits/audio sample (often padded out to 33 bytes/20 ms or 13.2 kbit/s). The quality of the coded speech is quite poor by modern standards, but at the time of development (early 1990s) it was a good compromise between computational complexity and quality, requiring only on the order of a million additions and multiplications per second. The codec is still widely used in networks around the world. Gradually FR will be replaced by Enhanced Full Rate (EFR) and Adaptive Multi-Rate (AMR) standards, which provide much higher speech quality with lower bit rate.

Technology

[edit]

GSM-FR is specified in ETSI 06.10 (ETS 300 961) and is based on RPE-LTP (Regular Pulse Excitation - Long Term Prediction) speech coding paradigm. Like many other linear predictive coding (LPC) speech codecs, linear prediction is used in the synthesis filter. However, unlike most modern speech codecs, the order of the linear prediction is only 8. In modern narrowband speech codecs the order is usually 10 and in wideband speech codecs the order is usually 16.

The speech encoder accepts 13 bit linear PCM at an 8 kHz sample rate. This can be direct from an analog-to-digital converter in a phone or computer, or converted from G.711 8-bit nonlinear A-law or μ-law PCM from the PSTN with a lookup table. In GSM, the encoded speech is passed to the channel encoder specified in GSM 05.03. In the receive direction, the inverse operations take place.

The codec operates on 160 sample frames that span 20 ms, so this is the minimum transcoder delay possible even with infinitely fast CPUs and zero network latency. The operational requirement is that the transcoder delay should be less than 30 ms. The transcoder delay is defined as the time interval between the instant a speech frame of 160 samples has been received at the encoder input and the instant the corresponding 160 reconstructed speech samples have been out-put by the speech decoder at an 8 kHz sample rate.[1]

Implementations

[edit]

The free libgsm codec can encode and decode GSM Full Rate audio.[2][3] "libgsm" was developed 1992–1994 by Jutta Degener and Carsten Bormann, then at Technische Universität Berlin.[4] Since a GSM speech frame is 32.5 bytes, this implementation also defined a 33-byte nibble-padded representation of a GSM frame (which, at a frame rate of 50/s, is the basis for the incorrect claim that the GSM bit rate is 13.2 kbit/s). This codec can also be compiled into Wine to provide GSM audio support.

There is also a Winamp plugin for raw GSM 06.10 based on the libgsm.[5][6]

The GSM 06.10 is also used in VoIP software, for example in Ekiga, QuteCom, Linphone, Asterisk (PBX), Ventrilo and others.

See also

[edit]

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Full Rate

View on Grokipedia
from Grokipedia
Full Rate (FR), also designated as GSM 06.10, is a digital speech coding standard developed by the European Telecommunications Standards Institute (ETSI) for the , the first widely deployed second-generation cellular network launched in 1991. It serves as the foundational for full-rate channels (TCH/F), compressing 20-millisecond frames of speech—consisting of 160 samples at an 8 kHz sampling rate in 13-bit uniform —into 260 bits at an average of 13 kbit/s, enabling efficient transmission while preserving intelligible speech suitable for mobile communications over limited bandwidth. The encoding process relies on Regular Pulse Excitation with Long Term Prediction (RPE-LTP), a technique that models the human vocal tract to predict and quantize speech signals, dividing each frame into four 5-millisecond subframes for parameter extraction including reflection coefficients, pitch lag, and excitation pulses. This standard, specified in ETSI's EN 300 961, includes both encoder and decoder functions with a theoretical minimum delay of 20 milliseconds, though practical implementations may add 3 to 8 milliseconds for processing, resulting in a total delay of 23 to 28 milliseconds, and incorporates features like codec homing sequences for bit-exact testing and compatibility with A-law or μ-law compressed inputs via the A-interface. As the initial for , Full Rate prioritized network capacity and compatibility, achieving a balance between quality and efficiency suitable for early mobile environments, though its speech intelligibility was later critiqued for artifacts in noisy conditions. It laid the groundwork for subsequent enhancements, such as the Enhanced Full Rate (EFR) codec introduced in 1995, which improved perceptual quality using (ACELP) at the same , and the Half Rate codec at 5.6 kbit/s for doubled at reduced fidelity. Despite the evolution to more advanced standards like Adaptive Multi-Rate (AMR) in third-generation networks, Full Rate remains relevant in legacy systems and certain embedded applications, with open-source implementations like libgsm facilitating its use in and archival audio processing.

History and Development

Origins in GSM Standardization

The was established in 1982 by the Conference of European Posts and Telecommunications administrations (CEPT) to develop a pan-European standard for digital mobile communications, driven by the need for a unified digital system to replace fragmented analog networks and facilitate cross-border in . This initiative addressed the inefficiencies of existing national systems, aiming for interoperability and efficient spectrum use in second-generation () cellular technology. The speech codec selection process, conducted under CEPT auspices from 1987 to 1988 and later overseen by the European Telecommunications Standards Institute (ETSI), involved evaluating multiple candidate algorithms to meet stringent requirements for speech quality, computational complexity, and robustness over noisy mobile channels. Six primary codec proposals were shortlisted for detailed testing in 1986, including submissions from (Regular Pulse Excitation-Linear Predictive Coding, or RPE-LPC), (Multi-Pulse Excitation-Long Term Prediction, or MPE-LTP), and representatives from , , , and the (sub-band coders). The RPE-LTP technique, a hybrid approach combining regular pulse excitation for the residual signal with long-term prediction to model pitch periodicity, was selected as the winner in 1988 for its optimal trade-off between natural-sounding speech reproduction and low processing demands suitable for early digital mobile hardware. Standardization efforts culminated in the publication of ETSI specification GSM 06.10 in 1990, which formally defined the Full Rate codec—based on the selected RPE-LTP method—as the foundational standard for Phase 1 deployment in networks. This document outlined the codec's integration into the GSM air interface, ensuring compatibility across European operators. Significant contributions to the codec's development came from industrial research at companies like Research Laboratories in , which advanced low-complexity analysis-by-synthesis techniques, alongside academic expertise from institutions such as the Technical University of Delft. These efforts focused on refining excitation models and prediction algorithms to achieve viable real-time performance on single-chip processors.

Adoption and Timeline

The first commercial GSM networks launched in 1991 in with the Radiolinja service operated by and Mobira, marking the initial deployment of the Full Rate speech as the default standard for voice transmission. This was followed by the rollout in in July 1992, where initiated its D1 network using the same . The Full Rate enabled efficient use of the 200 kHz carrier bandwidth through (TDMA), supporting up to eight simultaneous voice channels per carrier, which facilitated scalable network capacity in these early deployments. By the mid-1990s, subscriber numbers had surged past 10 million globally, with Full Rate serving as the foundational integral to the technology's rapid expansion. This growth propelled to dominate the market, achieving approximately 80% global by 2000 and underpinning the connection of hundreds of millions of users worldwide. The introduction of the Enhanced Full Rate (EFR) codec in 1995 by ETSI improved speech quality while maintaining compatibility with existing Full Rate infrastructure, beginning a gradual shift away from the original standard. This was further accelerated by the Adaptive Multi-Rate (AMR) codec's standardization in 1999 through ETSI and , which offered adaptive bit rates for better error resilience and efficiency, leading to Full Rate's replacement in most networks during the . By the , Full Rate usage had declined to legacy support primarily in developing regions where infrastructure persisted due to slower migration to higher generations. As of 2025, Full Rate sees minimal active deployment amid widespread , , and migrations, though it remains in some IoT applications and legacy systems reliant on for low-bandwidth connectivity. In the UK, operators like O2 plan to restrict services to IoT and emergency use starting in October 2026, with full national shutdowns targeted by 2033, signaling the codec's ongoing phase-out.

Technical Overview

Core Principles

The Full Rate (FR) speech codec, standardized for the (GSM), employs a hybrid coding approach that integrates (LPC) for short-term spectral modeling with long-term prediction (LTP) to capture the periodic components of speech signals. LPC analyzes the speech frame to derive filter coefficients that predict short-term correlations, effectively representing the spectral envelope, while LTP refines this by searching for the optimal pitch lag and quantizing the long-term prediction gain to capture periodicity in voiced speech. This combination allows the codec to efficiently model both the formant structure and pitch of human speech, achieving compression without excessive distortion. At the core of the excitation mechanism is Regular Pulse Excitation (RPE), which approximates the residual signal after LPC and LTP filtering by placing pulses on a fixed grid within sub-frames, rather than using an exhaustive search over all possible positions. This grid-based approximation significantly lowers while maintaining adequate representation of the excitation signal's distribution, making it suitable for the resource-constrained environments of early mobile networks. The RPE process involves downsampling the residual, selecting optimal grid positions and amplitudes, and quantizing them to form the codec's output parameters. Designed to meet the bandwidth limitations of GSM's digital , the FR targets toll-quality speech reproduction at a of 13 kbit/s, balancing audio fidelity, low latency (under 30 ms algorithmic delay), and modest processing demands compatible with 1990s hardware. Input consists of 13-bit linear (PCM) samples at an 8 kHz sampling rate, processed in 20 ms frames of 160 samples, while the output comprises 260 bits per frame encoding the LPC, LTP, and RPE parameters for transmission and subsequent synthesis. This framework ensured robust performance over noisy channels, prioritizing perceptual quality for conversational use.

Key Parameters

The Full Rate operates at a gross of 13 kbit/s, producing 260 bits per 20 ms frame, which corresponds to 1.625 bits per audio sample given the 8 kHz sampling rate. This rate encompasses all encoded parameters prior to channel coding for transmission. Speech frames consist of 160 samples over 20 ms, segmented into four subframes of 40 samples each to enable subframe-level processing of excitation and prediction parameters. This structure supports efficient analysis of speech dynamics within each 5 ms subframe. The employs an 8th-order analysis filter to model the short-term correlations in the speech signal, with reflection coefficients quantized using a total of 36 bits per frame. Long-term prediction uses lags ranging from 40 to 120 samples (5 to 15 ms), encoded with 7 bits per subframe, along with prediction gains quantized to 2 bits per subframe, for totals of 28 bits and 8 bits, respectively, across the frame. Regular pulse excitation generates the component using 13 evenly spaced pulses within each 40-sample subframe, with grid positions selected via 2 bits per subframe (choosing among four possible alignments, total 8 bits across the frame). The block amplitude is quantized logarithmically with 6 bits per subframe (total 24 bits), while the 13 pulse values are each represented using 3-bit adaptive after normalization, yielding 39 bits for the pulses per subframe (total 156 bits). The design prioritizes low latency, with end-to-end transcoder delay below 30 ms (theoretical minimum of 20 ms plus minimal processing overhead) and no look-ahead buffering required for frame encoding.

Encoding and Decoding Process

Linear Prediction Analysis

In the linear prediction analysis stage of the Full Rate codec, the input speech signal undergoes pre-emphasis using a high-pass filter to boost higher frequency components and compensate for the spectral tilt of the human vocal tract. The filter is defined by Hp(z)=1−βz−1H_p(z) = 1 - \beta z^{-1}, where β≈0.93\beta \approx 0.93 (precisely 28180×2−1528180 \times 2^{-15}), applied to the 13-bit linear PCM samples at an 8 kHz sampling rate. This is followed by windowing the pre-emphasized 160-sample frame (corresponding to 20 ms of speech) with a Hamming window to minimize during autocorrelation computation, ensuring accurate estimation of the short-term spectral envelope. The short-term predictor is modeled as an 8th-order all-pole filter, where the coefficients are derived using the method on the windowed signal. The function r(k)r(k) is computed for lags k=0k = 0 to 88 as r(k)=∑n=k159sw(n)sw(n−k)r(k) = \sum_{n=k}^{159} s_w(n) s_w(n-k), where sw(n)s_w(n) is the windowed pre-emphasized signal; this yields nine values for the 8th-order analysis. These values are then processed via the Levinson-Durbin (implemented as the Schur for ) to solve for the 8 reflection coefficients kik_i (for i=1i = 1 to 88), which lie in the range −1<ki<1-1 < k_i < 1 and represent the prediction error feedback in a lattice structure. The reflection coefficients are subsequently transformed into log-area ratios (LARs) using LARc(i)=12ln⁡(1+ki1−ki)\text{LAR}_c(i) = \frac{1}{2} \ln \left( \frac{1 + k_i}{1 - k_i} \right) to provide a perceptually uniform parameterization suitable for quantization. Quantization of the LPC parameters occurs once per frame on the LARs using scalar quantization with nonuniform codebooks tailored to each coefficient's dynamic range and perceptual importance, allocating a total of 36 bits: 6 bits each for LARs 1 and 2 (64 levels), 5 bits each for LARs 3 and 4 (32 levels), 4 bits each for LARs 5 and 6 (16 levels), and 3 bits each for LARs 7 and 8 (8 levels). This bit allocation reflects the higher sensitivity of lower-order coefficients to quantization error, with the quantized LARs (LAR_c) transmitted as part of the 260-bit frame. At the decoder, the received LAR_c values are interpolated between consecutive frames to generate subframe-specific LAR sets, which are inverse-transformed back to reflection coefficients for the short-term synthesis filter. The synthesis filter reconstructs the speech signal by inverse linear prediction, passing the excitation through an 8th-order lattice all-pole filter s^(n)=∑i=18ais^(n−i)+u(n)\hat{s}(n) = \sum_{i=1}^{8} a_i \hat{s}(n-i) + u(n), where aia_i are the derived predictor coefficients and u(n)u(n) is the decoded excitation; the output is then de-emphasized with the inverse pre-emphasis filter to recover the original spectral balance.

Regular Pulse Excitation

In the Full Rate codec, the Long Term Prediction (LTP) analysis models the periodic components of the speech signal within the LPC residual. For each 40-sample subframe, an adaptive search is performed to determine the pitch lag in the range of 40 to 120 samples and the associated gain that best predicts the current subframe from past residual samples. The selected lag is quantized to 7 bits per subframe (28 bits total per frame), and the gain to 2 bits per subframe (8 bits total), for a combined 36 bits for LTP parameters per frame. This periodic component is then subtracted from the LPC residual to yield the LTP residual, which captures the non-periodic excitation. The Regular Pulse Excitation (RPE) encoding further models the LTP residual by representing it as a sparse set of pulses on a coarse grid. The 40-sample LTP residual subframe is processed through adaptive sample rate decimation to select one of four possible grids (encoded with 2 bits per subframe), each defining 13 pulse positions spaced approximately every 3 samples. The amplitudes at these 13 positions are then quantized using 3 bits each via adaptive (APCM), allocating 39 bits per subframe for the pulses (156 bits total per frame). This approach efficiently captures the key energy concentrations in the residual while minimizing bit usage. To normalize the excitation signal, a separate gain scaling factor representing the subframe is quantized using 6 bits per subframe. This factor, derived from the maximum in the selected grid, scales the pulses during reconstruction to match the original residual's . In the decoder, the RPE pulses are first reconstructed by placing the quantized amplitudes at their grid positions within the 40-sample subframe, with zeros elsewhere, and scaling by the gain factor. The LTP contribution is added by shifting the previous frame's excitation by the lag and scaling by the gain, yielding the full excitation signal. This excitation is then passed through the LPC synthesis filter to produce the output speech samples, followed by post-processing such as de-emphasis.

Implementations

Software Libraries

The first open-source implementation of the Full Rate speech was libgsm, developed between 1992 and 1994 by Jutta Degener and Carsten Bormann at . This C-based library provides royalty-free encoding and decoding of GSM 06.10 audio, capable of real-time processing on low-end CPUs such as those from the early 1990s. The European Telecommunications Standards Institute (ETSI) released the official reference implementation as part of the GSM 06.10 specification, including fixed-point code for the RPE-LTP transcoder and test vectors for verification. This reference code serves as a baseline for compliant implementations and is freely available through ETSI documentation. Integrations of Full Rate support appear in multimedia libraries like FFmpeg, which uses libgsm for decoding GSM audio streams, and (Sound eXchange), which encodes and decodes the format via an external GSM library. Both tools enable media processing workflows, such as converting uncompressed audio to GSM-compressed files. Licensing for these libraries is permissive: libgsm follows a custom copyright allowing free use, modification, and distribution without royalties, while the ETSI reference code is freely available under ETSI terms. Original RPE-LTP patents for the codec expired around 2010, eliminating prior encumbrances and enabling widespread adoption post-expiry. Practical usage includes command-line tools; for example, libgsm's toast utility compresses WAV files to GSM format with toast input.wav output.gsm, while SoX supports similar conversions via sox input.wav -r 8000 -c 1 output.gsm. These tools facilitate audio archiving and legacy telephony applications.

Hardware and VoIP Applications

In the , the Full Rate codec was implemented on (DSP) chips from , such as the TMS320C5x series, which powered baseband processing in early GSM mobile phones. These DSPs handled real-time encoding and decoding of the RPE-LTP algorithm at rates up to 20 MIPS, enabling efficient speech compression within the power constraints of battery-operated devices. The Full Rate codec finds continued use in VoIP systems for interoperability with legacy GSM networks, particularly through gateways that bridge IP-based calls to cellular infrastructure. In the Asterisk PBX, support is provided via the built-in GSM codec module (format "gsm"), which implements the GSM 06.10 Full Rate standard to facilitate low-bandwidth audio transmission in hybrid VoIP-GSM setups. However, with global 2G network shutdowns accelerating in 2025 (e.g., T-Mobile in the US beginning February 2025), its use in live networks is declining, though software implementations persist for archival and simulation purposes. Despite the shift to advanced codecs, the Full Rate standard persists in certain legacy devices, including select satellite phones that leverage GSM-compatible audio processing for global coverage in remote areas. It is also embedded in amateur radio equipment for digital voice modes, where its low bitrate suits constrained bandwidth scenarios. On mobile platforms, Android and iOS apps such as VLC Media Player support playback of GSM Full Rate audio files (.gsm format), enabling decoding and reproduction of archived or legacy recordings. In modern niche applications, the Full Rate codec is employed in (SDR) projects, notably within , where encoder and decoder blocks simulate GSM systems for educational, research, and prototyping purposes. These implementations process 33-byte frames to generate 16-bit speech samples, facilitating full network emulation on SDR hardware like RTL-SDR or HackRF.

Comparisons and Successors

Versus Enhanced Full Rate

The Full Rate (FR) and Enhanced Full Rate (EFR) codecs both utilize the GSM full rate channel at a gross bit rate of 22.8 kbit/s, but differ in source coding rates and techniques: FR employs Regular Pulse Excitation with Long Term Prediction (RPE-LTP) at 13 kbit/s, while EFR uses Algebraic Code Excited Linear Prediction (ACELP) at 12.2 kbit/s. EFR delivers a marked improvement in speech quality over FR, attaining a Mean Opinion Score (MOS) of 4.1 in clean conditions and demonstrating superior degradation MOS (DMOS) under vehicle and street noise (4.25 and 4.18, respectively), compared to FR's MOS of 3.4 in clean speech and lower noise robustness scores (3.83 and 3.92 DMOS). This upgrade results in more natural-sounding speech and performance nearing wireline telephony standards like G.726 ADPCM. EFR incurs higher computational demands than FR due to its ACELP structure and larger excitation codebooks, requiring approximately 14-18 MIPS for encoder and decoder operations, roughly 3 to 5 times that of FR's optimized implementations around 1-5 MIPS. Introduced in 1995 by ETSI as an optional upgrade for Phase 2+, EFR maintains backward compatibility through network negotiation, allowing fallback to FR when is required.

Transition to Adaptive Multi-Rate

The Adaptive Multi-Rate (AMR) codec was standardized by the European Telecommunications Standards Institute (ETSI) in December 1999 under specification GSM 06.90, providing a variable bit rate speech coding algorithm operating at rates from 4.75 to 12.2 kbit/s across eight modes. It employs multi-rate Algebraic Code-Excited Linear Prediction (ACELP) to encode narrowband speech signals, enabling dynamic adaptation of the coding mode based on real-time assessments of radio channel quality via uplink and downlink quality indicators. This adaptability allows AMR to optimize speech quality in varying error-prone conditions, marking a significant evolution from the fixed-rate Full Rate (FR) codec. The primary motivations for transitioning from to AMR stemmed from FR's inefficiency in handling the noisy and error-prone mobile radio environments typical of networks, where its fixed 13 kbit/s rate offered limited robustness against channel impairments without flexibility for rate adjustment. AMR addressed this by supporting both full-rate (22.8 kbit/s gross) and half-rate (11.4 kbit/s gross) channel modes, allowing seamless switching to lower-rate modes during good channel conditions to double network voice capacity while maintaining or improving in adverse scenarios. As an interim measure, the Full Rate (EFR) had partially mitigated FR's shortcomings, but AMR's multi-mode design provided a more comprehensive solution for both and capacity demands. AMR's deployment accelerated with its adoption as the mandatory default speech in 3GPP Release 99, finalized in 2000 for systems, ensuring compatibility and widespread integration in both and emerging networks. By the mid-2000s, most operators had upgraded their infrastructure to support AMR, driven by its benefits in spectrum efficiency and the growing prevalence of compatible devices, which hastened FR's obsolescence in modern deployments. Despite AMR's dominance, persists for legacy support of pre-AMR mobile devices in remaining networks, ensuring during the phased transition to higher-generation technologies. However, with global sunsets planned to complete by around 2030 to reallocate for and , FR's role will ultimately be eliminated as these legacy networks are decommissioned. As of 2025, 131 networks are planned to shut down by 2030, with 61 within 2025 and 278 completed, planned, or in progress across 83 countries.

Performance and Quality

Computational Requirements

The GSM Full Rate codec was designed to operate in real-time on the limited processors (DSPs) available in early 1990s mobile phones, requiring approximately 1 MIPS for an optimized of both encoding and decoding processes. This low computational allowed it to fit within the constraints of battery-powered devices, with typical implementations achieving under 1 MIPS through structural optimizations in datapath design and . Memory requirements were modest, utilizing around 2 KB of RAM for dynamic buffers, state variables, and codebooks during processing of 20 ms speech frames at 8 kHz sampling. Fixed lookup tables, such as those for coefficients and excitation pulses, were stored in ROM, typically amounting to several kilobytes depending on the hardware platform. The codec's power efficiency was a key design goal for battery-constrained mobiles, with implementations consuming less than 50 mW at 5 V on contemporary DSP cores. On modern hardware, the Full Rate codec's demands are negligible.

Speech Quality Metrics

The speech quality of the Full Rate codec is commonly assessed using the (MOS), a subjective metric on a 5-point scale where 1 indicates poor quality and 5 excellent. In clean speech conditions with no errors, the MOS for Full Rate typically ranges from 3.6 to 3.8, reflecting acceptable but not toll-quality performance compared to uncompressed PCM references. Detailed subjective listening tests conducted by ETSI in the late 1980s and early 1990s, including evaluations of six candidate codecs, confirmed an average MOS of approximately 3.71 under error-free conditions (EP0). In the presence of , such as at 15 dB SNR or vehicular at 10 dB SNR, the MOS remains relatively robust at 3.83 to 3.92 under clean channel conditions, though perceived degradation occurs due to the codec's limited handling. Channel errors further degrade quality; for instance, at a carrier-to-interference ratio (C/I) of 7 dB (corresponding to error pattern ), the MOS drops to 2.73 in clean speech and 2.85 in vehicular , approaching fair-to-poor ratings. These results stem from ETSI and collaborative evaluations around 1988–1990, which emphasized subjective listening tests with diverse speakers and conditions to ensure robustness in mobile environments. Objective measures provide complementary insights into Full Rate's fidelity. Post-decoding signal-to-noise ratio (SNR) achieves approximately 20–25 dB in error-free conditions, as quantified by noise levels equivalent to 23–25 dB in ETSI tests relative to G.711 PCM benchmarks. Segmental SNR, which averages distortion over short frames (typically 20 ms), exceeds 20 dB for sine wave tests across 100–2000 Hz and is applied separately to voiced and unvoiced frames to capture variations in periodic and noise-like speech segments. These metrics highlight the codec's efficiency in modeling speech but reveal limitations in high-frequency preservation. A characteristic artifact of the Regular Pulse Excitation (RPE) method is a "buzziness" or rattle-like quality, particularly in voiced segments, arising from the fixed pulse grid and long-term prediction mismatches that introduce tonal described as hollow or metallic. Fricatives, such as /s/ and /z/ , exhibit notable due to spectral folding and from RPE down-sampling (decimation factor of 3), which inadequately regenerates high-frequency noise-like components around 2500–3500 Hz. Overall, Full Rate outperforms analog FM systems (e.g., superior to 14.7 dB SNR ) but falls short of toll quality, with intelligibility maintained at around 70–90% in moderate error conditions based on diagnostic rhyme tests from evaluations. Real-world channel tests showed graceful degradation, with acceptable performance (e.g., <13% class II errors) down to C/I of 7 dB, equivalent to bit error rates of several percent in environments.

References

  1. http://www.scholarpedia.org/article/Global_system_for_mobile_communications_%28GSM%29
  2. https://www.etsi.org/deliver/etsi_en/300900_300999/300961/06.00.00_40/en_300961v060000o.pdf
  3. https://vocal.com/speech-coders/gsm-fr/
  4. https://www.eetimes.com/sorting-through-gsm-codecs-a-tutorial/
  5. https://www.gsma.com/about-us/who-we-are/our-history/
  6. https://pure.tue.nl/ws/files/2416208/200512836.pdf
  7. https://www.electronics-notes.com/articles/history/cellphone-history/gsm-history-groupe-special-mobile.php
  8. https://www.ericsson.com/en/about-us/history/company/timeline-placeholder/timeline-1991-gsm-changes-everything
  9. https://www.telekom.com/en/media/media-information/archive/25-years-of-deutsche-telekom-ag-589922
  10. https://www.etsi.org/images/files/news/CreationOfStandardsForGlobalMobileCommunication.epub
  11. https://fortunly.com/articles/smartphone-market-share/
  12. https://www2.spsc.tugraz.at/people/franklyn/ICASSP97/pdf/scan/ic970771.pdf
  13. https://www.etsi.org/deliver/etsi_en/301700_301799/301703/07.00.02_60/en_301703v070002p.pdf
  14. https://www.1global.com/blog/iot/2g-and-3g-sunsetting-and-what-it-means-for-iot
  15. https://piblu.co.uk/2024/08/28/switch-off-announcement-o2-2g-and-3g/
  16. https://www.gl.com/voice-codecs.html
  17. https://abut.sdsu.edu/EE684/Chapter9.pdf
  18. https://www.quut.com/gsm/
  19. https://ffmpeg.org/ffmpeg-codecs.html
  20. https://manpages.ubuntu.com/manpages/jammy/man1/sox.1.html
  21. https://www.quut.com/berlin/gsm/COPYRIGHT
  22. https://www.ti.com/lit/an/spra033/spra033.pdf
  23. https://docs.asterisk.org/Operation/Asterisk-Audio-and-Video-Capabilities/
  24. https://www.t-mobile.com/support/coverage/t-mobile-network-evolution
  25. https://wiki.gnuradio.org/index.php/GSM_full-rate_Audio_Decoder
  26. https://www.etsi.org/deliver/etsi_en/300700_300799/300726/07.00.01_40/en_300726v070001o.pdf
  27. https://portal.etsi.org/stq/presentations/smg11.pdf
  28. https://vivonets.ece.ucsb.edu/casmagarticlefinal.pdf
  29. https://www2.spsc.tugraz.at/people/franklyn/ICASSP97/pdf/author/ic974085.pdf
  30. https://ieeexplore.ieee.org/document/609960
  31. https://www.etsi.org/deliver/etsi_tr/146000_146099/146076/16.00.00_60/tr_146076v160000p.pdf
  32. https://www.eurasip.org/Proceedings/Eusipco/Eusipco2000/SESSIONS/THUAM/SS1/CR1956.PDF
  33. https://www.3gpp.org/ftp/tsg_sa/tsg_sa/tsgs_22/docs/PDF/SP-030767.pdf
  34. https://www.electronics-notes.com/articles/connectivity/2g-gsm/audio-codecs-vocoders-amr-celp.php
  35. https://www.gsma.com/connectivity-for-good/spectrum/technology-upgrades-and-legacy-network-sunsets-on-the-rise/
  36. https://gsacom.com/paper/2g-3g-switch-off-july-2025/
  37. https://www2.spsc.tugraz.at/people/franklyn/ICASSP97/pdf/scan/ic974085.pdf
  38. https://www.adaptivedigital.com/gsm-fr/
  39. https://www.ti.com/lit/pdf/spra583
  40. https://ieeexplore.ieee.org/document/389629
  41. https://www.itu.int/dms_pubrec/itu-r/rec/m/R-REC-M.1309-0-199710-W!!PDF-E.pdf
  42. https://www.etsi.org/deliver/etsi_tr/146000_146099/146085/17.00.00_60/tr_146085v170000p.pdf
Add your contribution
Related Hubs
Contribute something
User Avatar
No comments yet.