WP1 Resilient Communications
Solutions will be researched to provide resilient communication, in the absence of a mine’s power and communication networks, capable of operating in non-ideal conditions such as through falls of rock. In particular, meshed VLF/LF through-rock radios and wire-guided radio will be investigated. Some of the output of this work package will take the form of stand-alone results (i.e. for person-to-person communication) while others will be used in the resilient survivable sensors (WP2) and unmanned exploratory vehicles (WP3).
The work is divided into the following tasks:
T1.1 – LF/VLF Propagation Characteristics
Modelling of underground propagation has been undertaken in previous RFCS projects, notably in ADEMA (coal seam propagation) and EMTECH. The work comprised a brief analysis of both a Large Immobile Surface Antenna (LISA) and a Grounded Horizontal Electric Dipole (G-HED). Such modelling has not, however, incorporated environmental data (e.g. background noise) or geology. This task will, therefore, supplement the modelling of the propagation from both magnetic and electric dipole antennas by collecting and using data on rock conductivity and background noise. In particular, the modelling will address, in detail, the benefit that is realised by operating in the transition zone, as opposed to within the near field; and will provide input into Task 1.3 for the converse situation where near-field operation would allow a degree of frequency re-use in a mesh network.
The modelling will use a software tool such as MatLab to produce a transfer function that can be analysed to allow the determination of the ideal operating point of the system. The capability for frequency re-use will be established by investigating the eye pattern at the receiver, in the presence of an interfering signal. Conducting media are highly dispersive; that is, they give rise to a phase distortion, which means that a simple measurement of propagation at one frequency is not sufficient to fully-characterise the medium because data pulses (comprising energy over a range of frequencies) are distorted. This problem will be assessed in the modelling, which will be extended to look at the effect of phase distortion on the bit-error rate of different modulation schemes.
T1.2 – LF/VLF Antenna Analysis and System Design
A number of magnetic dipole antenna designs – including induction loops, electrically short loops, ferrite rods, and more unusual and novel antennas, such as those made from amorphous metals – will be carefully modelled. Additionally, electric dipole antennas will be evaluated for comparison, including a novel antenna based around a ceramic disc, originally described in RFCS project ADEMA but not pursued experimentally. It was reported, in that project, that such an antenna was extremely difficult to drive efficiently, and this is also the case for many narrow-band reactive antennas. In more general terms it is the case that, in a non-radiating system such as this, it is necessary to consider all aspects of the system as a whole, including channel coding, amplifier design, power matching and antenna design. This task will investigate these points and answer questions including: (1) How is the range affected by the power, specific aperture, and bit rate; and how does this differ between a near-field device and one operating in the transition zone? (2) What is the relationship between bandwidth and power, and (3) Is there merit in considering ultra-wideband non-tuned antennas or field-gradient antennas (as an aid to noise cancellation)?
A theoretical analysis will be undertaken, studying a variety of antenna types (including induction loops, electrically short loops, rods, discs and grounded dipoles) and methods of driving them efficiently in wideband and narrowband modes. Next, development and characterisation of adaptive impedance-matching circuitry for these antennas will be undertaken. This will be especially important if adaptation of the transmission frequency is required in order to match the conditions in a particular mine – e.g. interfering signals, non-Gaussian noise and interference, characteristics of the rock, available node spacing and so on. Then, hardware will be developed for candidate antennas, selected as the potentially most suitable according to the modelling work. This part of the work will include the design and implementation of the radio front-end, modulation and demodulation, RF power amplifier and adaptive impedance-matching circuitry for the antenna selected, possibly including baseband digital processing. The antenna circuitry will be supplemented by basic software to allow streams of digital data to be transmitted with error detection monitoring and logging. Tests will be undertaken at underground locations in the UK and Spain where the quality of the digital link will be assessed.
T1.3 – LF/VLF Mesh Node Design
A through-rock LF/VLF mesh network will be developed to overcome the significant antenna and power requirements that would be typical for a low-frequency single-hop-link in coal-bearing geology. A salient point of the mesh network being proposed here is that the severe attenuation due to the skin-depth phenomenon is offset by the proximity of the nodes in the mesh. Recommendations on the distance and orientation of the nodes, and on the frequency and bandwidth, will rise out of T1.1 and T1.2. The mesh network development will be managed by structuring it as communications layers, the Physical Transmission layer being based on the through-rock propagation analysis in T1.1 and the design of the antenna and associated transmission and receiver front-end circuitry in T1.2. The present task, T1.3, adds the Network and Transport layers, with the aim of demonstrating the reliable transmission and acknowledgement of data segments between two points on the network across an arbitrary route, and presenting the whole design as a working prototype.
This will involve (1) building hardware to interface between a communicating device that presents a segment of data (i.e. an arbitrary sequence of bytes) for transmission, and the LF/VLF radio transmitter; (2) designing software to handle the networking and packet-assembly routines and (3) the converse operations at the receiver. Formalising the project into communications layers not only simplifies the work, but allows it to be more easily managed amongst the partners.
Initially, software tools will be developed to simulate a mine environment and to allow a variety of algorithms to be investigated on the desktop. It is anticipated that precise synchronization in the transmissions (wake-up and sleeping periods of nodes) will be needed in order to achieve maximum efficiency in the use of energy. Assessing the effects of de-syncing between nodes, signal level and signal-to-noise ratio, lost transmissions, transmission retries, etc. on the network’s effective bandwidth and on battery life (when operating on backup power supply) will require extensive computer simulations. Other factors, such as the network topology will also be included in the simulations.
Finally, equipment will be designed and prototyped to allow a simple network to be demonstrated in an underground location to provide person-to-person communication. A prototype will be produced to allow laboratory testing prior to field testing in WP6.
T1.4 – Hybrid Monofilar/Bifilar Mode Radio
Prior to the development of the leaky feeder communication systems that are used in mine galleries and transport tunnels, considerable work was carried out into monofilar and bifilar wire-guided radio. These will offer a major advantage over leaky feeder systems, for emergency temporary applications, because of the much reduced size and weight of the cable and hence increased speed and simplicity of deployment. Monofilar and bifilar modes each have their pros and cons so a hybrid system, that would provide the benefits of each, will be researched in this task to provide a system for person-to-person rescue communications.
In early work into these modes of communication, researchers noticed that if a 2-conductor cable was used, a degree of mode conversion between monofilar and bifilar occurs naturally, although the degree of the effect was not always sufficient to maximise performance. This was not adequately pursued because of the development of leaky feeder systems which were preferable for fixed (but not temporary) applications. It is the purpose of this task, therefore, to research methods of making mode conversion more reliable and determining the performance that such a system will be able to provide.
First, a 2-wire guidewire system will be modelled with the aim of quantifying the degree of natural mode conversion that occurs in various types of gallery and, accordingly, the performance of such a system, taking into account factors such as the lateral separation between handsets and the line and the longitudinal level of attenuation and hence of the range for a given power. Then, means of artificially improving and controlling the mode conversion will be researched and, again, the candidate systems will be modelled. The results of the modelling will be assessed in the light of the ergonomic considerations such as the requirement, or otherwise, to fit mode conversion units to the cable, the effect that any such mode convertors might have on the bulk and weight of the inherently compact cable, and the consequential differences in the ease of rapid deployment.
Next, an experimental verification of the modelling work will be carried out by trialling the most promising systems in galleries in mines with differing characteristics. Particular attention will be given to the longitudinal range and the degree to which communication handsets can be used without undue consideration to their proximity to the line, an important consideration for a “fit and forget” communications system and a major differentiator from previous wire-guided systems that have been developed for mines rescue purposes.
Finally, appropriate handsets will be prototyped for use as part of a wire-guided radio system laboratory tests conducted in preparation for field testing in WP6.
T1.5 – Common Communication Protocols
To ensure compatibility between the mesh network and the equipment that will utilise these communication systems (namely the resilient, survivable sensor to be developed in WP2 and the small exploratory vehicles to be developed in WP3), a set of common protocols will be standardised in this task. These will function somewhat like the Application Layer of common networks. That is, the nature of the packet data transmission, developed in T1.3 will be ‘invisible’ to these protocols. The common Application Protocols developed in this task will include a method of transmitting low-bandwidth encoded speech data, for use with the LF/VLF mesh network when used for person-to-person communications.
First, the software protocols will be defined followed by the hardware interface that will be used between the communications modules and the equipment that will utilise them. On the basis of the initial results from T1.1 and T1.2, the most suitable low-bandwidth speech coding algorithm for use with the LF/VLF mesh network will be determined. In addition, a suitable method of text transmission will be investigated as a backup, for person-to-person communication, when the characteristics of the communication channel dictate that speech is not achievable.