Overview
Voice communications has been the main service of radio networks since their inception. Single Sideband (SSB) voice is used in beyond-line-of-sight (BLOS) HF systems, while VHF systems traditionally use frequency modulation (FM). Practically, however, this analogue voice capability suffers three major drawbacks:
- Unsecure: Can be intercepted easily
- Unclear: Mainly due to channel noise
- Non-Specific: There is no way to select the other party
Digitising the voice communications allow these shortfalls to be overcome.
RapidM provides a best-in-class voice modem solution in the RA, RT and embedded product families, aimed specifically at HF beyond-line-of-sight (BLOS) communications. This next-generation solution is highly secure, significantly reduces operator fatigue and increases clarity and understandability both for point-to-point and broadcast voice communications. It allows for extended range superior voice communication performance even in adverse channel conditions.
The following products feature Secure Digital Voice (SDV) as a software function:
- RapidM Modem Modules when hosting EF-256 Embedded Functions for Secure Voice & Data
- RT7 Strategic Voice & Data Encryptor
- RT5 Tactical Voice & Data Terminal
- RT3 Tactical Voice & Data Encryptor
- RT1 In-line Voice & Position Encryptor
- RA7 Secure Voice & Data Modem
- RA5 Secure Voice & Data Terminal
- RA3 Secure Voice & Data Modem
- RA2 Secure Voice & Data Modem
- RA1 Secure Voice & Position Modem
Key Features
The RapidM Secure Digital Voice (SDV) is characterised by the following key aspects:
- Security: A strong modern digital encryption algorithm secures all communication.
- Clarity: Digital communications allow all channel noise to be removed, giving the user a clear, intelligible voice service which reduces operator fatigue.
- Private Line Capability: The caller can select a callee from an address book. Call capability utilises 3G ALE.
- Call alert: A ring indication alerts an operator of an incoming call.
- Automatic Data Rate Change: The Vocoder rates will automatically adjust between 600, 1200 and 2400bps to best suit the channel.
- Asymmetric Data Rates: Uplink and downlink Vocoder rates can be different to best suit the channels.
- Late Entry Resynchronisation: The SDV receiver can synchronise on an existing transmission even after missing the initial synchronisation. This resynchronisation capability is also very useful in deep slow fading HF channels.
- Robust Mode: The user can select 300bps rate. This requires slower speech and clear articulation.
- Language Independence: The Vocoders (especially *TWELP) provide high voice quality (PESQ scores) on a wide variety of global languages.
Solution Description
The RapidM Secure Digital Voice (SDV) modem solution tightly couples a dedicated digital voice waveform with strong encryption (AES-256) and the latest generation low-rate voice coder technology (*TWELP or **MELPe).
* TWELP voice coding by DSP Innovations Inc.(www.dspini.com)
** MELPe for US customers only
RapidM’s Secure Digital Voice solution can be used either one of the following two modes:
Broadcast: Initiated by simply pressing the press-to-talk (PTT) and talking, allowing everyone on the secure network to receive and decrypt the communications. Vocoder rates will start at 600bps and automatically change to 1200bps if the channel conditions allow. If PTT is pressed in a scanning network, a 3G broadcast link set-up (LSU) is first sent to all stations on the network.
Private Line: a call is established by selecting a recipient from an address book, allowing a secure voice connection between two parties. Calls can be established on automatically adjusted or fixed Vocoder rates varying between 300, 600, 1200 and 2400bps. Point-to-point calls are established using 3G ALE.
The 300bps robust secure voice rate requires the operators to speak slower and to clearly articulate. This ultra-low rate mode is intended for last ditch real-time voice communications between trained operators.
Automatic Data Rate Change (DRC) takes place by nodes continually exchanging received Link Quality Assessment (LQA) information, allowing the voice quality to be continually optimised for the channel condition.
Security
Because of its digital nature, RapidM Secure Digital Voice can be protected by Advanced Encryption Standard (AES) with 256-bit key length, which is a super strong trusted algorithm.
For a non-export-controlled alternative, Digital Encryption Standard (DES) with 56-bit key length can be used.
Dedicated RapidM Key Management Software (KMS) is used to create, wrap and transport unique, customer-defined cryptographic keys.
Export Control
Under the Wassenaar Arrangement, low rate voice coders are listed as munitions under export control for data rates below 700bps. Also under Wassenaar, symmetric encryption algorithms with key lengths greater than 56 bits are under export control.
| Service | Controlled | Non Controlled | |||
| Low-rate Vocoder | 300 | 600 | 700 | 1200 | 2400 |
| Crypto Algorithm | AES-256 | DES-56 & AES-256 | |||
Performance – Gaussian Channel
To evaluate the RapidM Secure Digital Voice solution, we compare several recordings made through the RapidM RS8 Channel Simulator using a Gaussian (AWGN) Channel, which is representative of a line-of-sight white noise environment. The results are graphed below as Intelligibility versus Signal-to-Noise Ratio (SNR). The intelligibility metric measures clarity relative to typical groundwave over-the-air HF channel conditions.
| # | Vocoder & Rate | Audio file | SNR (dB) | Channel | Graph |
| 1 | Original voice clip | >+30 | N/A | ||
| 2 | Over-the-air SDV signal | -3.0 | AWGN | ||
| 3 | Analogue Voice | -3.0 | AWGN | A | |
| 4 | 600bps TWELP | -3.0 | AWGN | A | |
| 5 | Analogue Voice | -1.5 | AWGN | B | |
| 6 | 600bps TWELP | -1.5 | AWGN | B | |
| 7 | Analogue Voice | 0 | AWGN | C | |
| 8 | 1200bps TWELP | 0 | AWGN | C | |
| 9 | Analogue Voice | +1.5 | AWGN | D | |
| 10 | 1200bps TWELP | +1.5 | AWGN | D | |
| 11 | Analogue Voice | +1.5 | AWGN | E | |
| 12 | 2400bps TWELP | +1.5 | AWGN | E | |
| 13 | Analogue Voice | +6.0 | AWGN | F | |
| 14 | 2400bps TWELP | +6.0 | AWGN | F | |
| 15 | Analogue Voice | -6.0 | AWGN | G | |
| 16 | 300bps TWELP | -6.0 | AWGN | G |
Performance – CCIR Poor Channel
To evaluate the RapidM Secure Digital Voice solution for a more typical beyond-line-of-sight (BLOS) sky-wave channel, we compare several recordings made through the RapidM RS8 Channel Simulator using a CCIR Poor channel. The results are graphed below as Intelligibility versus Signal-to-Noise Ratio (SNR). The intelligibility metric measures clarity relative to typical skywave over-the-air HF channel conditions.
| # | Vocoder & Rate | Audio file | SNR (dB) | Channel | Graph |
| 17 | Analogue Voice | +3.0 | CCIR Poor | H | |
| 18 | 600bps TWELP | +3.0 | CCIR Poor | H | |
| 19 | Analogue Voice | +6.0 | CCIR Poor | I | |
| 20 | 1200bps TWELP | +6.0 | CCIR Poor | I | |
| 21 | Analogue Voice | +8.0 | CCIR Poor | J | |
| 22 | 2400bps TWELP | +8.0 | CCIR Poor | J | |
| 23 | Analogue Voice | -0.5 | CCIR Poor | K | |
| 24 | 300bps TWELP | -0.5 | CCIR Poor | K |
Summary
It is evident that the RapidM Secure Digital Voice solution overcomes the shortcomings of analogue voice by providing a secure and clear digital voice communication service for tactical, naval, governmental and commercial users alike.