Some Preliminary Short-Range Transmission Loss Measurements For Wireless Sensors Deployed On Indoor Walls Konstantinos Sasloglou, Faisal Darbari, Ian A. Glover, Ivan Andonovic and Robert W. Stewart Institute for Communications & Signal Processing Department of Electronic and Electrical Engineering University of Strathclyde Glasgow, United Kingdom Email: ksasloglou@eee.strath.ac.uk Abstract—Antenna characteristics and propagation are of fun- damental importance to the coverage, capacity and service qual- ity of all wireless communication systems. This paper presents short-range narrowband propagation measurements at 2.445 GHz for sensor network applications in an indoor environment. The effect of sensor node location on a wall has been determined for a pair of linearly polarised rectaxial antennas and a pair of ceramic patch antennas. Propagation loss has been measured as a function of (i) node separation (i.e. link length), (ii) node drop (i.e. vertical displacement of nodes below the ceiling) and (iii) node height (i.e. the perpendicular displacement of the nodes from the wall surface). It is observed that there is no significant effect of wall offset. In addition, the path loss exponent n generally increases with decreasing node drop. I. I NTRODUCTION The physical channel plays a significant role in the design and deployment of wireless sensor networks which typically comprise a spatial distribution of low-power, low-cost, nodes. The performance and reliability of such a network is sensi- tively dependent on adjacent node transmission loss. Knowledge of transmission loss is required to establish appropriate node radiated power and/or spatial density. A consortium of five Scottish universities is, as part of a larger project (Specknet [1]), developing an 8 cm x 5 cm x 3 mm (i.e. credit-card sized), energy-neutral, wireless sensor node for deployment on flat surfaces in indoor environments. The nodes are powered by rechargeable solar cells (hence energy-neutral) and are therefore autonomous. They have low data-rate (less than 10 symbol/s). Applications for the resulting network include location tracking, environmental monitoring, fire and smoke detection etc. Since the nodes rely on light source for recharging solar cells it is required to know the optimum placement of these nodes on flat surfaces in order to receive maximum light besides giving minimum transmission loss between nodes. Measurements have been made to characterise transmission loss as a function of: • node separation (i.e. link length) • node drop (i.e. vertical displacement of nodes below the ceiling) • node height (i.e. the perpendicular displacement of the nodes from the wall surface) Many indoor measurements of transmission loss exist in the literature (e.g. [2], [3], [4], [5], [6]) These, however, primarily address the loss expected due to propagation through different building materials (comprising walls, ceilings, floors, doors etc.) and the expected transmission loss and fading statistics for propagation within room interiors rather than propagation over a room’s bounding plane surfaces. The study in this paper presents propagation loss measure- ments for indoor wireless sensors deployed on walls. II. METHODOLOGY Measurements of transmission loss have been made at 2.445 GHz using a pair of ceramic patch antennas and a pair of linearly polarised rectaxial antennas, Figure 1. Partial pattern information for these antennas is available in [7] and [8]. The transmitted signal was an unmodulated carrier. The return losses of the rectaxial and ceramic patch antennas were measured using a network analyzer and found to be better than -15 dB between 2.40 and 2.50 GHz [8] and better than -6 dB between 2.4 and 2.485 GHz respectively [7]. Their respective gains are 2.2 dBi and 2 dBi. Fig. 1. Rectaxial antenna (left) and ceramic patch antenna (right). All the measurements were made with the antennas flat against the wall surface. 1-4244-2424-5/08/$20.00 ©2008 IEEE ICCS 2008 129