Purpose: Use SWA sensor to measure treadmill walk and run as well as spot cycling sports energy consumption, and make comparison with indirect calorimetric testing results, make research on the effectiveness of SWA measurement to energy consumption in fitness, provide reference for sensor correct application. Method: 15 male adults complete 4.8 km/h、6.4 km/h、8.0km/h such 3 speeds walking and running in flat treadmill, and complete 10km/h、13km/L、 15km/h indoors spot cycling ,apply SWA( two software visions V6.1、V7.0) and indirect colorimetric method to predict energy consumption in exercising, make matched testing, regression analysis, correlation analysis, Le index as well as consistency check to prediction data. Results:(1) gWA6.1 overestimates energy consumption in treadmill walk and run( 1% overestimate)and spot cycle(9% overestimate),but get good correlation and consistency (r＝0.91， ICC＝0.91).（2）gara sensor predicted fitness energy consumption value far bigger than that by indirect calorimetric method, and the prediction difference would be increased as sports speed increasing, which is different from the sensor old version overestimates easy intensity sports energy consumption and underestimate hard intensity sport energy consumption, but energy consumption that predicted has high correlation and consistency with that by indirect calorimetric.