TERC SENSOR | A Non-Acoustical Sensor to Improve Battlefield Communication

If you're a soldier in a noisy tank or chopper, clear communication can mean the difference between life and death. That's why the military was looking for novel techniques to reduce the amount of background noise in high-noise environments like driving a tank or flying a helicopter.

A Non-Acoustical device will help troops communicate on a noisy battlefield. The Tuned Electromagnetic Resonance Collar -- or TERC sensor -- detects changes in the electrical field around the neck produced by moving vocal chords. A computer recreates the speech and sends it into the recipients' earpieces. The TERC sensor detects changes in the electrical field around the neck produced by moving vocal chords. This information, in the form of an electrical signal is reflected from the sensor, picked up by a separate device, and re-assembled into human speech.

WHAT IS THE TERC SENSOR?

The TERC is a non-acoustic voiced speech sensor that is worn like a collar around the neck of the subject. There are, in fact, several versions of the TERC sensor. The final version is shown in action below.The TERC sensor is made from a flexible plastic substrate, copper foil strips, two tunable inductors, two tunable capacitors, an SMA connector, and a fabric sleeve to facilitate hands-free operation. There are no batteries, switches, LEDs, or other components in the design.

HOW DOES THE TERC SENSOR WORK?

The basic priciple behind the TERC sensor is simple. The TERC sensor is essentially a big capacitor. The neck of the speaker forms the dielectric material inside the capacitor. When voiced speech occurs, glottal activity causes the dieletric properties inside of the capacitor to change slightly which, in turn, causes the capacitance to change slightly. Hence, the TERC sensor is essentially a capacitive method for detecting glottal activity.

Fig 1. Tuned Electromagnetic Resonator Collar (TERC) Sensor

The complicated part of the system is in detecting the very small changes in capacitance that occur during glottal activity. To do this, the TERC sensor also has a passive tunable matching network to create one or more resonant frequencies.

When the TERC sensor is excited by a low-power (typically -10dBm) sinusoidal waveform at its resonant frequency, very little of the forward power is reflected. When glottal activity occurs, small changes in the TERC's capacitance lead to slight shifts in the resonant frequency of the sensor. These slight shifts in resonant frequency lead to relatively large  changes in the amount of reflected power due to the steep slope of the sensor's resonance.

Hence, glottal activity can be measured by observing the reflected signal from the resonator. The reflected signal from the TERC sensor is obtained by using a device called a "circulator". Because the envelope of this reflected signal contains the desired speech information, it use a sensitive and sophisticated AM demodulator to recover the desired signal. Note that the TERC sensor does not apply any voltage or current to the neck of the speaker. The sinudoidal drive signal does establish a low-power electromagnetic field across the neck of the speaker.

SINGLE-MODE TERC SENSOR PROTOTYPE CONSTRUCTION

Fig 2. TERC sensor concept.


The single-mode TERC sensor is constructed on a 0.5 mm thick substrate of flexible clear acrylic plastic. Two appropriately sized strips are cut from a sheet of adhesivebacked 0.1 mm thick copper foil and affixed to the acrylic substrate.

The printed circuit board containing an SMA connector as well as a balanced passive matching network was then centred on the substrate and the appropriate connections are soldered to the copper foil strips. The printed circuit board used in the prototype is 1.57 mm thick and had traces printed on both the top and bottom layers. The SMA connector serves as the single connection from the sensor to the drive and demodulation circuitry.

Fig 3. Dimensions of the single-mode TERC sensor prototype and layout of the balanced passive matching network. All capacitors and inductors shown are tunable.

THE BALANCED

Matching network of tunable capacitors and inductors used to tune the resonance of the TERC sensor for each subject consisted of two Coil craft model 142-09J08 inductors with a tuning range of 0.47–0.57 μH and two Sprague–Goodman GCL40000 tunable plastic dielectric capacitors with a tuning range of 1–40 pF.

Two steps are also needed to increase the durability and comfort of the TERC sensor without compromising its function. Initial sensor prototypes occasionally experienced solder fractures at the connection between the rigid printed circuit board and the flexible foil strips. This problem was resolved by securing the printed circuit board to the sensor with a bead of electronic grade silicone rubber adhesive sealant. By applying the sealant to the entire perimeter of the printed circuit board, the durability of the prototype was significantly improved and no subsequent solder fractures were observed.

The comfort of the sensor is improved by enclosing the sensor in a nylon fabric sleeve with a Velcro closure. Holes are cut in the sleeve to allow for the coaxial cable connection to the SMA connector and also for tuning of the matching network. A photograph of the sensor in its sleeve as worn by a male subject is shown in Figure 4.

Fig 4. Photograph of the placement of the single-mode TERC sensor.

TERC SENSOR DRIVE AND DEMODULATION CIRCUITRY

Figure 5 shows an overview of the TERC sensor drive and demodulation circuitry. After the TERC sensor’s matching network has been tuned, and prior to the commencement of speech testing, the RF signal generator must be set to a frequency close to the TERC sensor’s resonant frequency.


Fig 5. Block diagram of the TERC sensor drive and demodulation systems.


The frequency of the drive signal is held constant over each test but may be adjusted slightly between experiments due to small changes in subject’s position and the consequent changes in the TERC sensor’s resonance. The 3 dB splitter and oscilloscope at the output of the circulator are used to facilitate these adjustments. The test technician finds the resonant frequency of the TERC sensor by adjusting the drive frequency until the amplitude of the reflected signal is minimized.

The circulator shown in Figure 5 is designed to provide unity transmission and high isolation over the expected range of TERC sensor resonant frequencies. Transmission is measured to be within ±0.2 dB and isolation is measured to be in excess of 25 dB between 20 MHz and 80 MHz. All ports of the circulator  are designed with a characteristic impedance of approximately 50 to avoid signal reflections.

THE ADVANTAGES OF TERC SENSOR

Non-acoustical sensors like TERC pick up only the sound of the person wearing the device. Conventional microphones pick up not just the speaker's voice, but also every other sound within its range, so there is a lot more background noise.

The TERC sensor is also highly immune to acoustic background noise. The TERC sensor can provide very clear information for voice activity detection and pitch estimation in high-noise environments. The TERC sensor is capable of detecting glottal activity that occurs during voiced speech as well as other glottal activity like swallowing.

The TERC sensor is not capable of detecting unvoiced speech like whispering, fricatives, and unvoiced plosives. These types of speech have no glottal activity and, since the TERC sensor is a glottal activity detector, are not measurable by the TERC sensor.

POSIBLE USES OF TERC SENSOR

Divers working underwater, crews in noisy vehicles, or soldiers on covert missions will use the sensors first. The technology is perfect for factory or constructor workers who must wear helmets for safety. But one day such sensors may allow people to use cell phones in places such as trains, theaters or libraries without disturbing those around them.

More Sensors

References:

DR Brown III, K Keenaghan and S Desimini: Measuring glottal activity during voiced speech using a tuned electromagnetic resonating collar sensor.

Kevin Michael Keenaghan: A Novel Non-Accoustic Voiced Speech Sensor | Experimental Results and Characterization.