Techniques for expeditiously predicting electromagnetic wave propagation

• US 5,623,429 A
• Filed: 04/06/1994
• Issued: 04/22/1997
• Est. Priority Date: 04/06/1994
• Status: Expired due to Fees

• Current Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Fortune, Steven J., Valenzuela, Reinaldo A.
• Primary Examiner(s): Teska, Kevin J.
• Assistant Examiner(s): LOPPNOW, MATTHEW CLAY
• Application Number:
  US08/223,630
• Time in Patent Office: 1,112 Days
• Field of Search: 364/578, 364/524, 455/67.6
• US Class Current: 703/3
• CPC Class Codes: G09B 23/181 for electric and magnetic fields; for voltages; for currents
• CPC Class Codes: H04B 17/3913 Predictive models, e.g. based on neural network models

Abstract

Techniques are disclosed for optimizing the prediction of RF propagation. A three-dimensional environment, such as a building, is modeled as a plurality of two-dimensional cross-sectional areas including a plurality of surfaces. Each surface is associated with a reflection coefficient and a transmission coefficient. A reference transmitter location and a plurality of reference receiver locations are selected. For each reference receiver location, RF propagation paths are determined with respect to the reference transmitter location. The reference transmitter location and the reference receiver locations represent propagation path endpoints. A plurality of parallel planes are used to partition the three-dimensional environment into a plurality of intervals. The RF reflective surfaces and propagation paths within each interval are projected into a cross-sectional area. A plurality of line segments are positioned in the cross-sectional area to form a plurality of triangular areas.

Claims

1. A method for optimizing the prediction of RF propagation in a three-dimensional environment comprising a plurality of surfaces, each surface having (i) a reflection coefficient specifying the ratio of reflected to incident RF energy, and (ii) a transmission coefficient specifying the ratio of RF energy transmitted through the surface to RF energy incident upon the surface, the method including the following steps:

(a) selecting a reference transmitter location within the three-dimensional environment;

(b) selecting a plurality of reference receiver locations within the three dimensional environment;

(c) for each reference receiver location, determining at least one RF propagation path between the reference receiver location and the reference transmitter location, the reference transmitter location representing a propagation path endpoint, and the reference receiver locations each representing propagation path endpoints;

(d) partitioning the three dimensional environment into a plurality of intervals along any one of the three dimensions using a plurality of parallel planes;

(e) for each interval, projecting the RF propagation paths and the surfaces within each interval into a two-dimensional cross-sectional area;

(f) inserting a plurality of line segments into each cross-sectional area to form a plurality of triangle walls, the plurality of triangle walls and the projected surfaces together forming a plurality of triangular areas;

(g) for each projected RF propagation path, tracing the projected RF propagation path through n triangular areas in the cross-sectional area by locating an nth triangular area selected from the plurality of triangular areas containing a first propagation path endpoint;

(h) for each of n triangular areas, determining the triangle wall intersecting the propagation path, and identifying an (n-1)th triangular area sharing this triangle wall;

(j) decrementing n by one, and recursively repeating the determination and identification steps set forth in (h) until a second propagation path endpoint is reached.

2. A method for optimizing the prediction of RF propagation in a three-dimensional environment as set forth in claim 1 wherein step (c) further includes the step of determining a plurality of RF propagation paths including a direct path joining the reference transmitter location to a given reference receiver location across a straight-line path, as well as one or more reflection paths joining the reference transmitter location to a given reference receiver location via reflections from one or more surfaces.

3. A method for optimizing the prediction of RF propagation as set forth in claim 2, wherein a plurality of reflection paths are determined such that all reflection paths having up to m reflections are considered, where m is an integer greater than zero.

4. A method for optimizing the prediction of RF propagation as set forth in claim 3 wherein each RF propagation path intersects a plurality of surfaces;

the method further including the step of calculating a propagation component setting forth propagation loss relative to free-space propagation as the product of the magnitude squared of the reflection coefficients and the transmission coefficients corresponding to the plurality of intersected surfaces.

5. A method for optimizing the prediction of RF propagation as set forth in claim 4 further including the step of calculating a local mean of received RF power at each Of the reference receiver locations as the scalar sum of all the propagation components reaching each of the reference receiver locations.

6. A method for predicting RF propagation in a three-dimensional environment including the following steps:

(a) selecting a reference transmitter location and at least one reference receiver location;

(b) determining a plurality of propagation paths between said reference transmitter location and each of said reference receiver locations;

said propagation paths including at least one direct path joining said reference transmitter location to one of said reference receiver locations along a straight-line path, and at least one reflection path joining said reference transmitter location to one of said reference receiver locations via at least one surface;

(c) associating each of said surfaces with a reflection coefficient specifying the ratio of reflected to incident RF energy, and a transmission coefficient specifying the ratio of RF energy transmitted through the surface to incident RF energy;

(d) projecting the three-dimensional environment and the plurality of propagation paths into a plurality of two-dimensional cross-sectional areas;

(e) partitioning the two-dimensional cross-sectional areas into a plurality of triangular regions;

(f) for each propagation path, calculating a propagation path component representing propagation loss relative to free-space propagation of RF energy from a reference transmitter at the reference transmitter location producing a reference RF power level, the propagation loss being equal to the product of the magnitude squared of the reflection and transmission coefficients; and

(g) for each reference receiver location, calculating a local mean of received power equal to the scalar sum of the powers of all of the propagation path components corresponding to the reference receiver location.

7. The method of claim 6, wherein the propagation component of each reflection path including one reflective surface is calculated using a ray imaging procedure comprising the following steps:

(a) from the at least one reference receiver location, tracing the reflection path backwards, in a direction opposite to the direction of travel of the RF energy, to encounter a first reflective surface;

(b) determining a first reference receiver location image about the first reflective surface as the reference receiver location reflected about a plane including the first reflective surface; and

(c) calculating the overall reflection path length as the distance between the reference transmitter point and the first reference receiver location image.

8. The method of claim 6 Wherein the propagation component of each reflection path including m surfaces, m being greater than 1, is calculated using a ray imaging procedure comprising the following steps:

(a) from the at least one reference receiver location, tracing the reflection path backwards, in a direction opposite to the direction of travel of the RF energy, to encounter an mth surface;

(b) determining an mth reference receiver location image about the mth surface as the reference receiver location reflected about a plane including the mth surface;

(c) from the mth surface, tracing the reflection path backwards to encounter an (m-1)th surface;

(d) determining an (m-1)th image about the (m-1)th surface as the nth reference receiver location image reflected about a plane including the (m-1)th surface;

(e) repeating steps (c) and (d) recursively until the reflection path is traced back to the reference transmitter location; and

(f) calculating the overall reflection path length as the distance between the reference transmitter point and the last reference receiver location image determined in step (e).

9. A method for predicting RF propagation in a three-dimensional environment including the following steps:

(a) selecting a reference transmitter location and at least one reference receiver location from within the three-dimensional environment;

the three-dimensional environment comprising a plurality of surfaces, each surface having (i) a reflection coefficient specifying the ratio of reflected to incident RF energy, and (ii) a transmission coefficient specifying the ratio of RF energy transmitted through the surface to RF energy incident upon the surface;

(b) determining a plurality of propagation paths between said reference transmitter location and each of said reference receiver locations;

said propagation paths including at least one direct path joining said reference transmitter location to one of said reference receiver locations along a straight-line path, and at least one reflection path joining said reference transmitter location to one of said reference receiver locations via at least one surface;

the reference transmitter location representing a propagation path endpoint, and the reference receiver locations each representing propagation path endpoints;

(c) partitioning the three dimensional environment into a plurality of intervals along any one of the three dimensions using a plurality of parallel planes;

(d) for each interval, projecting the RF propagation paths and the surfaces within each interval into a two-dimensional cross-sectional area;

(e) inserting a plurality of line segments into each cross-sectional area to form a plurality of triangle walls, the plurality of triangle walls and the projected surfaces together forming a plurality of triangles;

(f) for each projected RF propagation path, tracing the projected RF propagation path through n triangles in the cross-sectional area by locating an nth triangle selected from the plurality of triangles containing a first propagation path endpoint;

(g) for each of n triangles, determining the triangle wall intersecting the propagation path, and identifying an (n-1)th triangle sharing this triangle wall;

(h) decrementing n by one, and recursively repeating the determination and identification steps set forth in (g) until a second propagation path endpoint is reached;

(j) for each propagation path, calculating a propagation path component representing propagation loss relative to free-space propagation of RF energy from a reference transmitter at the reference transmitter location producing a reference RF power level, the propagation loss being equal to the product of the magnitude squared of the reflection coefficients and the transmission coefficients associated with the surfaces of the propagation path; and

(k) for each reference receiver location, calculating a local mean of received power equal to the scalar sum of the powers of all of the propagation path components corresponding to the reference receiver location.

10. A method for predicting RF propagation in a three-dimensional environment including the following steps:

(a) selecting a reference transmitter location and at least one reference receiver location;

(b) determining a plurality of propagation paths between said reference transmitter location and each of said reference receiver locations;

said propagation paths including at least one direct path joining said reference transmitter location to one of said reference receiver locations along a straight-line path, and at least one reflection path joining said reference transmitter location to one of said reference receiver locations via at least one reflective surface;

(c) associating each of said reflective surfaces with a reflection coefficient specifying the ratio of reflected to incident RF energy, and a transmission coefficient specifying the ratio of RF energy transmitted through the surface to incident RF energy;

(d) partitioning the three dimensional environment into a plurality of intervals along any one of the three dimensions using a plurality of parallel planes;

(e) for each interval, projecting the RF propagation paths and the surfaces within each interval into a two-dimensional cross-sectional area;

(f) inserting a plurality of line segments into each cross-sectional area to form a plurality of triangle walls, the plurality of triangle walls and the projected surfaces together forming a plurality of triangles;

(g) for each projected RF propagation path, tracing the projected RF propagation path through n triangles in the cross-sectional area by locating an nth triangle selected from the plurality of triangles containing a first propagation path endpoint;

(h) for each of n triangles, determining the triangle wall intersecting the propagation path, and identifying an (n-1)th triangle sharing this triangle wall;

(j) decrementing n by one, and recursively repeating the determination and identification steps set forth in (h) until a second propagation path endpoint is reached;

(k) for each propagation path, calculating a propagation path component representing propagation loss relative to free-space propagation of RF energy from a reference transmitter at the reference transmitter location producing a reference RF power level, the propagation loss being equal to the product of the magnitude squared of the reflection and transmission coefficients; and

(m) for each reference receiver location, calculating a local mean of received power equal to the scalar sum of the powers of all of the propagation path components corresponding to the reference receiver location.